Limiting factors and thresholds for macroinvertebrate assemblages in European rivers: Empirical evidence from three datasets on water quality, catchment urbanization, and river restoration

It has been widely stated that pressures acting at large spatial scales influence local habitat conditions and might limit the effects of local restoration measures. However, only a few empirical studies have used statistical methods that have explicitly been developed to investigate such wedge-shaped relationships. The objectives of the present study were (i) to identify pressures acting as limiting factors and to investigate the mitigating effects of local restoration measures in three datasets from European rivers, (ii) to derive thresholds for the ecological status of invertebrates, and (iii) to compare results derived from two statistical approaches, one using aggregated response variables like biological metrics (quantile regression trees), another using taxon-specific responses to derive separate community thresholds for the negative response of sensitive and the positive response of tolerant taxa (Threshold Indicator Taxa Analysis, TITAN). The results indicated that wedge-shaped relationships, typically resulting from limiting factors, are common in datasets from Central European rivers. There was empirical evidence for limiting effects of water pollution and catchment land use and an indication of a mitigating effect of hydromorphological restoration measures. The results emphasize the need to consider such large-scale pressures in river management and restoration because they potentially constrain the effects of local restoration measures. The thresholds derived for the aggregated response variables (metrics) and the community thresholds for sensitive taxa were in good agreement with values reported in the literature but differed markedly depending on the statistical method used. A possible reason is the different focuses of the methods on (i) the threshold for an aggregated response variable (metric), which includes the negative and positive response, and hence, reflects ecosystem functioning, and (ii) the community threshold of sensitive taxa based on taxon-specific negative responses, which is possibly best suited for species conservation issues. However, this interpretation requires further analysis since the results of the two methods showed no consistent differences. (C) 2011 Elsevier Ltd. All rights reserved.